See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/301565036

Best options for the exposure of traditional and innovative moss bags: A
systematic evaluation in three European countries
Data · April 2016

CITATIONS READS
3 193
15 authors, including:

Fiore Capozzi Simonetta Giordano

University of Naples Federico II University of Naples Federico II

25 PUBLICATIONS 207 CITATIONS 113 PUBLICATIONS 2,489 CITATIONS
SEE PROFILE SEE PROFILE

Jesús Aboal Paola Adamo

University of Santiago de Compostela University of Naples Federico II

132 PUBLICATIONS 2,505 CITATIONS 136 PUBLICATIONS 3,834 CITATIONS
SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Evo-devo View project

BEMOFA - Sondierung für die Entwicklung von moosbewachsenen Gebäudefassadenpaneelen View project

All content following this page was uploaded by Teresa Boquete on 22 April 2016.

The user has requested enhancement of the downloaded file.

 Environmental Pollution 214 (2016) 362e373

Contents lists available at ScienceDirect

Environmental Pollution
journalhomepage:www.elsevier.com/locate/envpol

Best options for the exposure of traditional and innovative moss bags:
*
A systematic evaluation in three European countries
a, h a, * b c d b
F. Capozzi , S. Giordano , J.R. Aboal , P. Adamo , R. Bargagli , T. Boquete ,
c b e, f a g h i
A. Di Palma , C. Real , R. Reski , V. Spagnuolo , K. Steinbauer , M. Tretiach , Z. Varela ,
j b
H. Zechmeister , J.A. Fernandez
a Dipartimento di Biologia, Universita di Napoli Federico II, Campus Monte S. Angelo, Via Cinthia 4, 80126 Napoli, Italy
b Department of Cellular Biology and Ecology, Faculty of Biology, University of Santiago de Compostela, 15782 Santiago de Compostela, Spain
c Dipartimento di Agraria, Universita di Napoli Federico II, Via Universita, 100, 80055 Portici (NA), Italy
d Dipartimento di Scienze Fisiche, della Terra e dell’Ambiente, Universita di Siena, Via P.A. Mattioli 4, 53100 Siena, Italy
e Plant Biotechnology, Faculty of Biology, University of Freiburg, Schaenzlestr. 1, 79104 Freiburg, Germany
f BIOSS e Centre for Biological Signalling Studies, 79104 Freiburg, Germany
g GLORIA-Coordination, Institute for Interdisciplinary Mountain Research, Austrian Academy of Sciences & Center for Global Change and Sustainability, University of
Natural Resources and Life Sciences Vienna, Austria
h Dipartimento di Scienze della Vita, Universita di Trieste, Via L. Giorgieri 10, 34127 Trieste, Italy
i BIOVIA Consultor Ambiental, Edificio Emprendia, Campus Vida, 15782 Santiago de Compostela, Spain
j Dept. Botany and Biodiversity Research, University of Vienna, Rennweg 14, 1030 Vienna, Austria

article info abstract

Article history: To develop an internationally standardized protocol for the moss bag technique application, the research team participating in
Received 24 February 2016 the FP7 European project “MOSSclone” focused on the optimization of the moss bags exposure in terms of bag characteristics
Received in revised form (shape of the bags, mesh size, weight/surface ratio), duration and height of exposure by comparing traditional moss bags to a
1 April 2016
new concept bag, “Mossphere”. In particular, the effects of each variable on the metal uptake from the air were evaluated by a
Accepted 12 April 2016
systematic experimental design carried out in urban, industrial, agricultural and background areas of three Euro-pean countries
with oceanic, Mediterranean and continental climate. The results evidenced that the shape, the mesh size of the bags and the
exposure height (in the tested ranges), did not significantly influence the uptake capacity of the transplanted moss. The aspects
Keywords:
more affecting the element uptake were represented by the density of the moss inside the bags and the relative ratio between
Active biomonitoring
Air pollution its weight and the surface area of the bag. We found that, the lower the density, the higher the uptake recorded. Moreover,
Moss uptake three weeks of exposure were not enough to have a consistent uptake signal in all the en-vironments tested, thus we suggest an
Trace elements exposure period not shorter than 6 weeks, which is appropriate in most situations. The above results were confirmed in all the
Mossphere countries and scenarios tested. The adoption of a shared exposure protocol by the research community is strongly
Pseudoscleropodium purum recommended since it is a key aspect to make biomonitoring surveys directly comparable, also in view of its recognition as a
monitoring method by the EU legislation.

© 2016 Elsevier Ltd. All rights reserved.

1. Introduction anthropogenic sources. Even though evidence of specific compo-nents of this

mixture to drive major risk for human health remained for long inconclusive,
Outdoor air pollutants are a complex mixture of primary and secondary the particulate matter (PM) has recently been designed as a Group 1
compounds originating from a myriad of natural and carcinogen by the International Agency for Research on Cancer (IARC). Fine
airborne particles contain metals, polycyclic aromatic hydrocarbons (PAHs)
and other toxic chemicals and can increase the natural-cause mortality even at
* concentra-tions well below the European annual mean limit value (Beelen et
This paper has been recommended for acceptance by Hageman Kimberly Jill.
*
Corresponding author. al., 2014). Even if some metals in (wet, dry and occult)
E-mail address: giordano@unina.it (S. Giordano).

http://dx.doi.org/10.1016/j.envpol.2016.04.043
0269-7491/© 2016 Elsevier Ltd. All rights reserved.
 F. Capozzi et al. / Environmental Pollution 214 (2016) 362e373
atmospheric deposition are among potentially toxic pollutants, in Europe air
quality target values have been established only for As, Cd, Ni and Pb and
most automatic air quality monitoring stations measure the concentrations of
particles based on size (aero-dynamic diameter of PM10: 10 mm or PM2.5:
2.5 mm) and not their chemical composition. Moreover, when some data on
metal deposition are available they have poor spatial coverage and local
sources or variations in their fluxes remain hidden. Thus, to obtain
quantitative information on the spatial pattern of deposition of metals
(especially those not measured by monitoring devices) the monitoring with
suitable organisms has become a very common approach (e.g., Harmens et al.,
2015 and references therein). The moss bag technique is probably the most
applied method for the active monitoring of airborne trace elements in urban
and indus-trial environments. This technique was introduced by Goodman and
Roberts (1971), later modified by Little and Martin (1974) and during the last
decades there were several investigations pointing to the optimization and
standardization of the method (Gailey and Lloyd, 1986a, b, c, d; Ares et al.,
2012, 2014; Giordano et al., 2013). However, most studies considered only
one of few methodological steps of the moss bag technique such as the
duration or the height of bag exposure and were developed in areas with
specific climatic and environmental conditions (e.g. Ares et al., 2012).
Taking advantage of the FP7 European project “MOSSclone” focused on
the culture of a particularly performing moss clone, the production of a new
concept bag (“Mossphere”) for the moss exposure and the development of a
standardized protocol for the moss bag technique, the research team involved
in the project undertook a complex and systematic evaluation of the most
important variables affecting the results of metal biomonitoring with moss
bags. In particular, the effects of each variable (1. Shape of bags; 2. Net mesh
size, 3. Ratio between moss weight and bag surface area; 4. Duration of the
exposure, 5. Height of the exposure) on the metal uptake were evaluated
separately. To develop an internationally standardized protocol for the moss
bag technique it seemed necessary to test the variability in each of the
methodo-logical steps in a range of climatic conditions and land use classes.
To this end an experimental design was, for the first time, applied in urban,
industrial, agricultural and background areas of three Eu-ropean countries
with oceanic, Mediterranean and continental climate.
2. Material and methods
2.1. Preparation of moss transplants
Moss transplants were prepared with Pseudoscleropodium purum (Hedw.)
M. Fleisch., one of the most commonly used species in the moss bag
technique (Ares et al., 2012). Samples were collected in a background area of
SE Galicia (NW Spain; X: 596060,
Y: 4709910 UTM 29N ETRS89) selected on the basis of previous results
(Boquete et al., 2013).
All moss samples were carefully separated from litter and other
exogenous materials and air-dried in laboratory. For the experi-ments, only 5
cm long green apices were selected. This material was firstly cleaned by
placing in a plastic sieve (0.7 cm mesh size) and then subjected to one wash
of 20 min with 10 mM EDTA (12.5 g d.w. moss/1 L EDTA with shaking) and
three washes of 20 min each with distilled water (10 g d.w./1 L distilled water
with shaking) and washed 3 times for 10 min in bidistilled water (10 g d.w./1
L bidistilled water with shaking) to remove adhering soil particles. The
samples were then blotted on filter paper to remove excess moisture.
Afterwards, moss apices were devitalized following three consecutive drying
cycles of 8 h each at 50 C, 80 C and 100 C (modified from: Adamo et al.,
2007; Giordano et al., 2009). Finally,
363
the bags were prepared (see below), vacuum packed and stored until use.
2.2. Experimental set-up
All the field experiments were carried out in NW Spain, SW Italy and E
Austria (Fig. 1; Tables S1, S2, S3). The climate in Galicia (NW Spain) is
influenced by the ocean and is temperate maritime; high rainfall (1000e2000
mm per year) and mild temperatures (annual average, 13 C and spring
average, 15 C) characterize the investi-gated area (www.meteogalicia.es). The
climate in Campania (SW Italy) is mild and influenced by the sea, (annual
average tempera-ture, 10.5 C and spring average, 13.5 C). The annual rainfall
ranges between 900 and 1200 mm (http://www.sito.regione.campania.it/ ).
Austria is not bordering the sea and presenting a temperate/ continental
climate in the investigated area (annual rainfall 550e900 mm; annual average
temperature, 11.3 C and spring average, 6.5 C) (http://www.zamg.ac.at/).
In each country seven exposure sites (ESs) affected by different level and
types of contamination were selected and classified accordingly as
agricultural, background, industrial, and urban sites. The bags were hung
vertically from sticks of an inert material fixed perpendicularly to a pole, or
similar structures, at a height of 4 m above the ground, except in experiment
2.2.5. (see below). The moss bags were exposed for 3 weeks, except in
experiment 2.2.4. (see below). Three replicates per ES for every single
treatment were exposed. Ten moss bags, vacuum-packed in polyethylene bags
and stored at 4 C, were used to check contamination after exposure during
transportation and laboratory handling.
2.2.1. Shape of the bags
Two couplets of moss-bags of different shape were compared at parity of
mesh size (2 mm), quantity of devitalized moss filled in, and external surface
of the device: rounded bag vs. Mossphere (S 30) e both made with a dry
2
mass/surface ratio of 30 mg cm e and flat bag vs. Mossphere (S15) e both
2
made with a dry mass/surface ratio of 15 mg cm -, Fig. 2.
The Mossphere is a device designed by our team consisting of two coaxial
empty spheres, each formed by two hemispheres, made of pierced high-
density polyethylene (the internal sphere), and of a 2 mm mesh nylon net)
(the external sphere). The internal sphere is 10 cm in diameter and has 3 mm
long spikes homoge-neously distributed on the convex side. The external
sphere is 11 cm in diameter. The two spheres are closed with four plastic
wires passing through four holes in the equatorial plastic border that delimits
each hemisphere. The space between the two spheres (10 mm thick) is filled
with moss, which is maintained in place by the spikes of the inner sphere.
The rounded bags were made as described by Ares et al. (2014). A square
of plastic net of 22 22 cm was filled with the moss material, and secured with
a nylon thread.
2
Rectangular flat bags (approximately 700 cm ) were made with plastic net
(2 mm mesh size). The moss was distributed homoge-neously inside the bag,
and to minimize overlapping and compression of the moss during the
exposure (hanging vertically) (Temple et al., 1981), the bag was sewn in a
zig-zag pattern with nylon thread to make 3 compartments.
Prior to use, the plastic net was washed in HNO 3 and then in distilled
water to eliminate any trace contaminants. Transplants were exposed in
triplicate for three weeks in all the ESs in March 2013; total number of
samples ¼ 189 (3 shapes 3 countries 7 ESs 3 replicates).
2.2.2. Mesh size
Mosspheres with different mesh size (1 mm, 2 mm, and 4 mm)
364 F. Capozzi et al. / Environmental Pollution 214 (2016) 362e373

Fig. 1. Maps showing the seven stations distributed in industrial, urban, agricultural and background scenarios in Spain (A), Italy (B) and Austria (C).

Fig. 2. Different shapes of moss bags tested. a) Rounded; b) Flat; c) Mossphere.

were filled with 11.40 g of dry moss material (dry mass/surface ratio: 30 mg
2
cm ) and exposed in triplicate in all the ESs at 4 m above the ground for
three weeks in March 2013; total number of samples ¼ 189 (3 mesh size 3
countries 7 ESs 3 replicates).
2.2.3. Ratio between moss weight and bag surface area
In order to investigate the effect of weight, Mosspheres with a nylon mesh
of 2 mm were filled with 5.70, 11.40 or 17.10 g d.w. of moss material, in order
2 2
to have weight/surface ratios of 15 mg cm (W15), 30 mg cm (W30) and 45
2 2
mg cm (W45), respectively. The Mosspheres were exposed in triplicate in
all the ESs at 4 m above the ground for three weeks in March 2013; total
number of samples ¼ 189 (3 wt/surface ratios 3 countries 7 ESs 3 replicates).
2.2.4. Duration of exposure
For this assay, Mosspheres with a dry mass/surface ratio of
2
30 mg cm and 2 mm mesh were exposed in triplicate at 4 m above the
ground. Three different durations of exposure (3, 6 and
12 weeks) were tested in parallel in all the ESs between March and June
2013, so there were in total 4 subsequent exposure periods of 3 weeks (n ¼
252; 3 countries 7 EEs 4 periods 3 replicates), 2 subsequent exposure periods
of 6 weeks (n ¼ 126; 3 countries 7 EEs 2 periods 3 replicates) and 1 exposure
period of 12 weeks (n ¼ 63; 3 countries 7 EEs 3 replicates).
2.2.5. Height of exposure
For this assay, Mosspheres with a dry mass/surface ratio of
30 mg cm and 2 mm mesh were exposed in triplicate at 4, 7 and 10 m above
the ground in 5 ESs of each country, excluding the agricultural and
background ones (n ¼ 135; 3 countries 5 ESs 3 heights 3 replicates). The
experiment was carried out in a single period of three weeks between March
and April 2013.
2.3. Sample preparation and chemical analysis
At the end of the exposure period, the mosses were dried
at 40 C until constant weight. The moss tissue was separately
 F. Capozzi et al. / Environmental Pollution 214 (2016) 362e373
homogenized from each bag (Total No. of bag processed: 1143), in heavy
metal-free mills (Retsch ZM 200 and Retsch PM100). Moss sub-samples of
0.5 g were then digested in 1 mL H 2O2 (30%, Sigma Aldrich) and 5 mL aqua
regia (1 HNO3:3 HCl) in a microwave (CEM Mars 5) and then filtered.
Concentrations of metals and metalloids included in the EU directives (As,
Cd, Hg, Ni and Pb), as well as indicators of industrial (Al, Ba, Be, Cr, Co, Cu,
Se, Sn, Sr, V and Zn) and traffic (Pd, Pt and Rh) emissions were determined
by inductively coupled plasma mass spectrometry (ICP-MS e Varian 820-
MS) at TE Labs (Tullow, Ireland). Mercury was determined in an elemental
analyzer (Milestone DMA 80). Digestion solutions without mosses (one
solution every 10 samples) were used as blanks and analysed in parallel. To
control the analytical quality, analytical replicates were processed, 1 every 10
samples and the standard deviation of analytical replicates was calculated.
Certified reference material (M3 Pleurozium schreberi; Steinnes et al., 1997)
was analysed in parallel, 1 every 10 samples. Contamination during
processing was controlled for by the use of analytical blanks (1 every 10
samples analysed).
Recovery of elements from the reference materials ranged be-tween 88%
for Ba to 119% for Ni. The relative standard deviation (RSD) was not higher
than 17% (Cr), except for As (50%). The overall error (calculated as e [(jX
Vj)/V]*100 -, where X is the mean concentration detected by ICP-MS and V
the certified value of M3) associated with the analytical process was usually
lower than 8% and never higher than 19%, with the only exception for As
(90%). The concentrations of Be, Co, Pd, Pt, Rh and Sn were under detection
limits in the reference material.
2.4. Data analysis
The limit of quantification of the technique (LOQT) was calcu-lated from
the initial concentrations as follows: xCi þ 1.96sCi, where xCi is the mean
value of the initial concentration in unex-posed moss samples (n ¼ 10) for
each element determined, and sCi is the corresponding standard deviation
(Couto et al., 2004 as modified in Ares et al., 2015). Calculation of the LOQT
enables clear distinction of the concentrations in exposed and unexposed
moss, which is also subject to different sources of variability. This should not
be confused with the LOQ of the analytical process with analytical standards.
Comparisons between the different exposure options were made separately
for each country and were based on those elements showing concentrations
higher than the LOQT at least in the arbitrarily fixed limit of 60% of the whole
dataset. All comparisons were carried out by using a non parametric Wilcoxon
matched pairs test (for 2 groups, also used as a post-hoc test when H0 was
refused with Friedman ANOVA test) or Friedman (for 3 or more groups) tests
by STATISTICA and R software. The ratios be-tween median absolute
deviation (MAD) and median were calcu-lated to evaluate the data spread of
each solution tested. The selection of the best solution was done on the basis
of the effects the various tested options had on the moss uptake and on the
replicability of the results.
3. Results and Discussion
3.1. Chemical analysis
As there were no significant differences between the final con-centration
of elements in the control moss bags and the initial concentration in
unexposed moss, we concluded that no contam-ination occurred during moss
transportation and handling in the laboratory.
The elements useful for our comparison were Al, Ba, Cr, Cu, Fe, Hg, Ni,
Pb, Sr and Zn; all the others were below the detection limit.
365
This outcome was not due to the analytical method applied since we obtained
good percentages of recovery (section 2.3). Probably, in the chosen sites,
some elements were present at low concen-trations and the exposure period
(section 3.5) was too short. Moreover, the three countries were characterized
by different pollution levels: Italy > Austria > Spain (according to the EEA re-
ports “Air quality in Europe” 2012e2013), consequently only those elements
complying with the criteria explained above were considered (section 2.4).
3.2. Shape of the bags
According to the data analysis criteria described in the section 2.4, for the
comparison between F (flat) and S15 (Mossphere) it was possible to use all the
elements except Cr and Hg, while for the comparison between R (rounded
bag) and S30 Mossphere, only Al, Ba, Cu, Ni, Sr (only in Italy) and Zn were
useful for our purposes.
3.2.1. Flat vs. S15 mossphere
Fig. 3 shows the comparisons for this assay, separately for each country,
associated to the statistical significance. In Austria, Al and Zn were found
mostly accumulated in the moss exposed in the flat bags, while that placed in
S15 had a higher amount of Ni and Pb. In Italy, a higher signal for Al and Zn
was recorded in flat bags, while S15 allowed a better uptake for Ba, Cu, Fe,
Ni, Pb and Sr. In Spain, the concentrations of Al, Ba, Cu, Ni, Pb and Zn were
higher in the flat bags, no differences in terms of enrichment were observed
for the remaining elements. A general trend is clear for the enrichment of Al
and Zn, for which the flat bags allowed a higher uptake perfor-mance. For the
remaining elements, we found uneven outcomes, suggesting that local
situations (e.g. weather conditions, orienta-tion of the flat bags) affected the
uptake performance more than the shape per se. In terms of data spread, no
clear differences were evidenced by the MAD/median ratio (Table 1).
Gailey and Lloyd (1986d) reported a better accumulation by the moss
Hypnum sp. exposed in rounded bags with respect to the flat bags; although
this is not evident in our results, we support the use of (sub-)spherical bags
since they allow a uniform collection effi-ciency from all space directions.
The flat bags in which two di-mensions are prevalent on the third one, besides
being a handmade product, have the problem that the moss uptake can depend
on the exposure orientation especially in our set-up because we fixed the bag
with two cable ties enabling only one degree of freedom (i.e. rotation around
its holder). If the pollutants come from a specific direction, the orientation of a
flat bag may largely affect the con-centrations found in the moss. Besides, the
upper and the lower sides of the flat bags are consistently smaller than the
other two, reducing the collection of the particulate matter falling from the top
or re-suspended from the soil. Anyway, no one of the two tested options
seems to be the best in all situations, but both are able to discriminate among
the different scenarios (see paragraph 3.7).
3.2.2. Rounded vs. S30 mossphere
In Austria, the moss exposed in the rounded bags had the highest
concentration of Ba, Cu and Zn. In Italy, the rounded bags ensured a better
uptake for Ni and Zn, while Al and Sr where higher in the moss contained in
S30. In Spain, the enrichment of Al, Cu, Ni and Zn was higher in the rounded
bags, while no differences were found for the remaining elements (Fig. 3).
There is a clear general trend only for Zn, for which the rounded bag always
allowed the highest uptake. No clear differences in terms of data spread were
evidenced by the MAD/median ratio (Table 1). This comparison did not yield
a clear result; the two options behaved quite similarly, likely for the same
density of the moss material inside the bag (see paragraph 3.4). Probably the
higher moss density, joint to the
366 F. Capozzi et al. / Environmental Pollution 214 (2016) 362e373

1
Fig. 3. Box-plots of element concentrations (mg g ) in the moss exposed in bags of different shape in Austria, Italy and Spain (flat ¼ F vs S15 Mossphere, grey; rounded ¼ R vs S30 Mossphere,
white). The dashed line represents the LOQ T. BOX: inside band ¼ median; extremities ¼ 1st and 3rd quartiles; whiskers ¼ MIN and MAX. W is the value of Wilcoxon test. “þ” indicates the only
option(s) fulfilling our criterion; *p < 0.05, **p < 0.01, ***p < 0.001.

Table 1 LOQT and therefore were not used in our evaluations; this was true for all the
Mean values of the ratio between median absolute deviation (MAD) and median obtained by countries (see Fig. 4). In Austria, the elements complying with our criterion
using “moss-bags” of different shape (flat ¼ F vs S15 mossphere; rounded ¼ R vs S30 were Ba, Cu, Fe, Hg and Zn. Only for Ba and Zn the 4 mm mesh seemed to
mossphere) exposed in Austria, Italy and Spain. Values are expressed as percentage.
ensure a better uptake performance, whereas no significant differences were
observed between the tested mesh sizes for Cu, Fe and Hg. In Italy, only Al,
Country Shape Al Ba Cr Cu Fe Hg Ni Pb Sr Zn
Ba, Cr, Fe and Sr were above the LOQ T. The 1 mm mesh appeared to
Austria F 13 7 5 7 8 3 17 5 2 13
S enhance the uptake of Cr, while the 4 mm mesh was the best option for Ba
15 11 4 23 12 9 2 20 16 2 28
R 6 1 16 12 8 2 23 8 3 13
and Sr. In Spain, the 1 mm mesh was the best option for the detection of Cu
S
30 6 4 4 5 4 3 19 1 1 12 and Fe.
Italy F 8 4 16 3 6 3 19 7 3 9
S 3 1 12 7 5 2 9 8 1 17
15 Overall, the three mesh sizes differed statistically only for a few elements.
R 5 3 29 5 4 2 15 8 3 8 In some cases the 4 mm mesh seemed to allow a better performance of the
S 11 2 13 3 2 4 16 2 2 12
30

Spain F 14 4 27 8 7 3 14 14 2 12
moss material with respect to the other two mesh sizes, but this result was not
S 5 6 17 3 5 3 15 3 2 20 always confirmed. The few cases in which the 4 mm size appeared to be the
15

R 11 5 7 6 4 3 21 11 2 8 best option could be explained by the loss of material (approximately 20%)

S 1 4 8 4 4 3 10 5 3 22
30 from the bags that, as a consequence, caused a reduction of the weight/bag
sur-face ratio and, hence, increased the uptake (section 3.4). No clear
differences in data spread were found for different mesh sizes (Table 2) This
exposure time (3 weeks), resulted in a reduced number of elements useful for finding is in agreement with Giordano et al. (2013) who evidenced, for the
comparison. Zopf., the mesh capability to
lichen Pseudevernia furfuracea
homogenise and reduce the variation in element enrichment
3.3. Mesh size and this seemed true independently of the mesh size interval
employed. According to different studies, the choice of an
Post-exposure values of Ni and Pb never exceeded the respective
 F. Capozzi et al. / Environmental Pollution 214 (2016) 362e373 367

1
Fig. 4. Box-plots of element concentrations ( mg g ) in the moss exposed in Mosspheres with different mesh sizes (1, 2, 4 mm) in Austria, Italy and Spain. The grey dashed line represents the LOQT.
BOX: inside band ¼ median; extremities ¼ 1st and 3rd quartiles; whiskers ¼ MIN and MAX. W, Q are the values of Wilcoxon and Friedman ANOVA tests. “þ” indicates the only option(s) fulfilling
our criterion; *p < 0.05, **p < 0.01, ***p < 0.001.

Table 2 conclude that mesh size does not affect the uptake of elements in moss in a
Mean values of the ratio between median absolute deviation (MAD) and median obtained by significant way; the selection of the proper mesh size must thus take into
using Mosspheres with different mesh sizes (1, 2, 4 mm) exposed in Austria, Italy and Spain. account only the loss of material during the exposure.
Values are expressed as percentage.

Country Mesh size (mm) Al Ba Cr Cu Fe Hg Ni Pb Sr Zn

Austria 1 3 4 9 8 6 6 11 5 3 13
2 6 4 4 5 4 3 19 1 1 12 3.4. Ratio between moss weight and bag surface area
4 5 3 17 7 6 4 18 3 3 17
Italy 1 6 2 21 4 5 5 11 6 3 6
It was not possible to carry out the statistical analysis for Cr, as most of
2 7 2 24 5 2 2 15 5 2 16
4 7 2 24 5 2 2 15 5 2 16 the data did not satisfy the 60% criterion. For all the remaining elements, the
Spain 1 8 8 13 10 12 2 24 2 4 13 final concentrations increased in the Mossphere filled with the smallest
2 1 4 8 4 4 3 10 5 3 22
amount of moss. The W15 was the only option fulfilling the LOQT for most
4 6 2 5 6 5 3 13 1 3 15
of the elements as the case of: Ba, Fe, Ni and Pb in Austria; Pb and Ni in
Italy; Cu, Ni and Sr in Spain. In all the other circumstances, when W30 and
W45 also fulfilled the criterion, W15 significantly differed from the other two
inadequate mesh size may lead to the loss of large amounts of material, setups giving always the “largest signal” (Fig. 5). In replicability terms no
principally due to weather conditions (e.g. Archibold, 1985; Strachan and clear trend was evidenced by the ratio between median absolute deviation
Glooschenko, 1988; Ares et al., 2012). We can (MAD) and median (Table 3). Our results reflect
368 F. Capozzi et al. / Environmental Pollution 214 (2016) 362e373

1
Fig. 5. Box-plots of the element concentrations (mg g ) in the moss exposed in Mosspheres prepared with different moss weight/sphere surface ratio (W15, W30 and W45) in Austria, Italy and
Spain, see text for specifications. The dashed line represents the LOQ T. BOX: inside band ¼ median; extremities ¼ 1st and 3rd quartiles; whiskers ¼ MIN and MAX. W, Q are the values of Wilcoxon
and Friedman ANOVA tests. “þ” indicates the only option(s) fulfilling our criterion; *p < 0.05, **p < 0.01, ***p < 0.001.

Table 3 Sphagnum denticulatum Brid., observed a general increase in the moss post-
Mean values of the ratio between median absolute deviation (MAD) and median obtained by exposure concentrations of Cd, Pb and Zn when they decreased the moss
2
using Mosspheres with different moss weight/sphere surface ratio (15, 30 and 45 mg cm ) amount inside the bags.
exposed in Austria, Italy and Spain. Values are expressed as percentage.
2
Moreover the smallest ratio we applied (15 mg cm ), did not affect the
Country Weight Al Ba Cr Cu Fe Hg Ni Pb Sr Zn
data spread in a significant way, since it was already adequate to minimize
replicability problems. Indeed, this ratio is 3 times higher than that suggested
Austria W15 11 4 23 12 9 2 20 12 2 28
by Ares et al. (2014) as a compromise between pollution signal and data
W30 6 4 4 5 4 3 19 2 1 12
replicability. Therefore, we conclude that the quantity of moss contained
W45 2 2 13 9 12 1 16 6 3 14
Italy W15 3 1 12 7 5 2 9 8 1 17 within the Mossphere significantly affects the uptake performance of the
W30 11 2 13 3 2 4 16 2 2 12 device; in particular, the smaller the amount of moss, the better uptake
W45 3 0 11 6 4 1 18 14 2 14 performance, with no effect on the data spread, particularly when the ratio is
Spain W15 5 6 17 3 5 3 15 3 2 20 2
not higher than 15 mg cm . This is probably the result of a more
W30 1 4 8 4 4 3 10 5 3 22
W45 5 1 12 5 5 2 12 4 2 12 homogenized distribution of elements on the surface of the receptor (moss),
thereby minimizing analytical variation.

in part those of Zechmeister et al. (2006) and Ares et al. (2014): the maximum
element interception is gained when the moss material is exposed to the air in
a thin layer with all the shoots equally exposed, with no or scarce overlap 3.5. Exposure time
among gametophores and without leaflets flattening. In particular, Ares et al.
(2014), by using In this experiment all the studied elements showed values
 F. Capozzi et al. / Environmental Pollution 214 (2016) 362e373 369

above the LOQT at least for some countries and exposure times (Fig. 6). Table 4
Aluminium, Ba and Pb were the only elements for which no differences were Mean values of the ratio between median absolute deviation (MAD) and median obtained by
using Mosspheres exposed for 3, 6 and 12 weeks in Austria, Italy and Spain. Values are
observed in any country or treatment. In Austria the highest uptake rates ( mg expressed as percentage.
1
g per week) corresponded to 3 weeks for Cu and 12 weeks for Hg. In Italy
Time (weeks) Al Ba Cr Cu Fe Pb Hg Ni Sr Zn
Cu, Fe, Ni and Sr showed the highest rates after 6 weeks, and Cr and Hg after
Austria 3 6 4 8 8 6 7 3 20 4 17
12 weeks. In Spain the 3-week exposure period caused the highest uptake
6 8 3 17 4 9 5 2 17 5 13
rates for Ni and Sr but in the case of Fe and Zn it was obtained after 12 12 9 8 13 7 9 9 2 15 3 12
weeks. Overall, exposure periods of 6 or 12 weeks seem, without great Italy 3 6 3 20 3 6 4 2 17 2 10
differences between them, to enhance uptake rates for a higher number of 6 5 2 24 8 5 3 1 30 2 9
elements. Nevertheless, the 12 week period was the only one showing the 12 6 3 28 3 4 7 2 11 4 12
highest uptake rates for at least one element in all the countries. As in the Spain 3 5 3 17 6 6 9 2 11 2 15
previous experiments, no clear differences in terms of data spread were 6 8 5 16 4 6 15 2 21 2 12
12 7 11 22 6 6 4 4 14 2 18
evidenced by the MAD/median ratio (Table 4).

In the moss bag technique the duration of the exposure period has been
one of the most studied aspects (e.g. Gailey and Lloyd, 1986c; Tavares and
Vasconcelos, 1996; Adamo et al., 2003; Basile et al., 2008; Anicic et al.,
2008; Giordano et al., 2009; Ares et al., 2014). In general, the evaluation of
the time effect was based on
the accumulated concentration of pollutants in the moss and its associated
variability; but in a thoroughly review of the literature, in which at least three
different exposure times were tested (Ratcliffe, 1975; Gailey and Lloyd,
1986c; Tavares and Vasconcelos, 1996; Vasconcelos and Tavares, 1998;
Basile et al., 2008, 2009;

1
Fig. 6. Box-plots of the accumulation rate (mg g per week) in the moss exposed in Mosspheres for 3, 6 and 12 weeks in Austria, Italy and Spain, see text for speci fications. BOX: inside band ¼
median; extremities ¼ 1st and 3rd quartiles; whiskers ¼ MIN and MAX. W, Q are the values of Wilcoxon and Friedman ANOVA tests. White boxes: values below the reference criterion; grey boxes:
values above the criterion; “þ” indicates the only option(s) fulfilling the criterion; *p < 0.05, **p < 0.01, ***p < 0.001.
370 F. Capozzi et al. / Environmental Pollution 214 (2016) 362e373
Anicic et al., 2009), Ares et al. (2012) suggest to assess the effect of this
variable in terms of uptake rates. These authors stated that the uptake rate
rarely depend on time. Similar results have been recently reported by Ares et
al. (2014) after testing the effect of 4-, 8- and 12 week exposure periods in the
uptake rates of Cd, Cu, Hg, Pb and Zn in Sphagnum denticulatum bags
exposed to different levels of pollution in Galicia (NW Spain). These authors
pointed out that the uptake rate tended to be temporally stable, independently
of the duration of the exposure period.
Our results agree with this trend: Al, Ba and Pb did not show differences
in uptake rates in any country, whereas in the case of Cr, Hg and Zn only one
or two exceptions (out of a total of 10 cases) were identified for the 12-weeks
period (i.e. Cr in Italy, Hg in Austria and Italy, and Zn in Spain). All the other
elements showed variable results, e.g. higher uptake rates of Cu for the 3
weeks period in Austria compared to 12 weeks in Italy and the same goes for
Sr after 6 weeks in Italy and Spain. However, in general terms our results do
not show a relationship between exposure periods and uptake rates. Hence, it
can be concluded that the best option is to expose the moss bags for a period
not shorter than 6 weeks because it ensures an adequate signal in terms of
enrichment of pollutants. Variations in metal uptake are likely a result of the
deposition mode (dry, wet or occult) and therefore, the longer the exposure,
the more homogenized these variations are. Neverthe-less, when mosses are
exposed in the surroundings of pollution sources, shorter exposure periods
could be enough.
3.6. Effect of the exposure height
It was not possible to carry out the statistical analysis for Cr and Hg, as
most of the data did not satisfy the 60% criterion (Fig. 7). The 7 m height was
the only option fulfilling LOQT for Al, and Pb in Spain, and the 4 m for Pb in
Austria. For Ba, Cu, Ni and Sr it was possible to test the hypothesis in Italy
and Spain, without finding supporting results; the same occurred for Fe in
Spain. The only element which reached LOQ T in all cases was Zn; for this
element no differences were found in Austria, while significant differences
were found in Italy (10 m height) and in Spain (4 m height), being always
higher at lower heights. No clear differences in terms of data spread were
evidenced by the MAD/median ratio (Table 5).
The vertical profile of contamination is the result of the complex process
of dispersion of contaminants in air masses and surface deposition, which is
affected by factors related to atmospheric conditions or the type of
distribution of contaminants (i.e. associ-ation with particles of different sizes
or in gas phase), as well as aspects such as air turbulence, specific location of
the source of emission (i.e. the vehicle exhaust pipes e fumes e and wear and
tear of wheels and asphalt e particles and the emissions from do-mestic fuel)
or topography of the area (i.e. presence of walls, buildings) (Anicic et al.,
2009; Adamo et al., 2011; Vukovic et al., 2013). As a result of these processes
some authors have found vertical profiles in the transplant concentrations.
Adamo et al. (2011) found that Hypnum cupressiforme bags exposed at a
height of 4 m in a street canyon in Naples were more efficient at retaining
contaminants associated to traffic and suspended dust (Al, As, Ba, Co, Fe, Pb,
Ti, V, and Zn) than samples exposed at a height of 20 m, which captured
contaminants associated with long distance transport and cations of marine
origin (i.e. Cr, K, Mg, and Mn). The same was found by Vukovic et al. (2013)
in Belgrade using Sphagnum girgensohnii bags with higher concentrations of
Al, Ba, Co, Cr, Cu, Ni, Pb, Sr, V and Zn at lower exposure height (i.e. 4 m)
than at higher heights (i.e. 8 and 12 m). These results are consistent to those
obtained for Zn in Italy and Spain and for Pb in Austria and Spain in this
study. Both elements are related to traffic emissions and abrasion processes
(Laschober et al., 2004; Zechmeister et al.,
2005; Napier et al., 2008; Thorpe and Harrison, 2008).
However, as for the other elements determined in the present work, most
authors did not find any vertical patterns of moss bags concentrations in
previous studies. Rivera et al. (2011) did not find any differences between Al,
As, Cd, Cr, Cu, Mo, Pb, Sb, Sn and Zb concentrations in Hylocomium
splendens bags exposed on balconies at heights of 3e21 m in Girona (NE
Spain). De Nicola et al. (2013) concluded that there were no differences in
Cd, Cr, Cu, Fe, K, Mg, Mn, Na, Ni, Pb and Zn concentrations in Hypnum
cupressiforme transplants among samples exposed at heights of 3, 6 and 9 m
in street canyons in Naples. In the same way, Vukovic et al. (2013) did not
find differences for Ca, Cd, Fe, K, Mg, Mn and Na concentrations in
Sphagnum girgenshonii transplants exposed at 4, 8 and 16 m in Belgrade
(Serbia). Finally, Ares et al. (2014) did not find differences in concentrations
of Cd, Cu, Hg, Pb and Zn in Sphagnum dentic-ulatum moss bags exposed at
0.5, 1, 1.5, 2, 2.5, 3, 4 and 5 m height in Galicia (NW Spain); although in
several instances highest concen-trations of all elements were yielded at 5 m
height, with the exception of a busy roadside, where the highest
concentrations were found at a height of 0.5 m.
According to Ares et al. (2014) the results are highly variable and
different contamination processes may be captured at different heights;
nevertheless, a specific height must be established to standardize this aspect of
the moss bags technique. Taking into account practical considerations (e.g.
assessment of the quantities of contaminants inhaled by people from the air
and avoiding vandalism), an exposure height of 4 m is recommended.
3.7. Evaluations of different scenarios (urban, industrial,
agricultural, background)
Element contents in the post-exposed mosses, regardless of shape and
mesh-size, indicated the industrial as the most impacted sites followed by the
urban and agricultural, down to the back-ground sites. No one of the above
options provided an unambigu-ous signal, so we conclude that shape and
mesh did not have an important influence on the uptake performance (see
sections 3.2 and 3.3) in each scenario. However, this was not true for the
weight/surface ratio, which, according to this experiment, should be not
2
higher than 15 mg cm to allow a better uptake, and hence a clear
discrimination among exposure sites (section 3.4).
It is worth noting that in the industrial scenarios a lower number of
differences among the tested times was recorded. This likely depends on the
higher pollution level characterizing these sites, thereby allowing a higher
uptake rate also after shorter exposure periods. Whereas in the background
areas, where lower pollution levels occur, the longest exposure period was
always needed to achieve sufficient element enrichment in moss.
The duration of exposure indicated that the 12 week option assured the
higher uptake of Cr, Hg and Zn in all scenarios. The same solution was also
the best option for Fe and Ni in urban, Ni in agricultural, Al in industrial, Ba
and Pb in the background sites. By the way, in this latter scenario, 6 and 12
week exposure allowed a similar enrichment of Al, Fe and Ni. No differences
among the up-take rates of Cu and Sr occurred among the three tested
periods, and this was true in all scenarios.
Exposure lasting 6 weeks seems long enough to detect
atmo-spheric trace element occurrence in moderately to
highly polluted areas. A longer exposure time might be
necessary in clean areas, or for those elements whose
concentration in the air is typically low. It should also be
taken into account that an exposure longer than 6 weeks, can
result in enhanced enrichment of a restricted set of elements,
but could also increase the probability of a loss of other
elements due to leaching and adverse meteorological
conditions (e.g. washing out by rain), even at parity of
exposure time, as
 F. Capozzi et al. / Environmental Pollution 214 (2016) 362e373 371

1
Fig. 7. Box-plots of element concentrations (mg g ) in the moss exposed in Mosspheres at 4, 7 and 10 m above the ground in Austria, Italy and Spain. The grey dashed line represents the LOQT.
BOX: inside band ¼ median; extremities ¼ 1st and 3rd quartiles; whiskers ¼ MIN and MAX. Q are the values of Friedman ANOVA test. “þ” indicates the only option(s) fulfilling our criterion; *p <
0.05, **p < 0.01, ***p < 0.001.

Table 5 reached the threshold. For the remaining comparisons, no differ-ences in

Mean values of the ratio between median absolute deviation (MAD) and median obtained by terms of uptake were found among the mosses exposed at different heights.
using Mosspheres exposed at different heights (4, 7 and 10 m) in Austria, Italy and Spain.
These outcomes further confirm the absence of any vertical gradient of the
Values are expressed as percentage.
elemental concentrations in mosses hanged between 4 and 10 m above the
Height (m) Al Ba Cr Cu Fe Pb Hg Ni Sr Zn ground in the study at hand.
Austria 4 2 3 4 6 6 3 1 15 5 23
7 12 5 19 7 3 3 2 7 6 6
10 2 3 15 8 4 0 1 11 5 10
4. Conclusions
Italy 4 3 2 6 3 2 0 1 10 1 9
7 1 3 6 4 6 0 2 23 3 26
10 5 3 5 4 2 0 4 14 0 16 The optimization for the exposure conditions, aim of the present study,
Spain 4 5 6 13 2 4 5 2 10 3 12 took into account some aspects already tested by other au-thors and reviewed
7 12 0 8 4 4 5 2 17 2 8 by Ares et al. (2012, 2014); the uniqueness of this work is represented by the
10 4 0 10 3 3 0 1 13 1 6
systematic experimental design applied, which allowed to test separately the
variables possibly affecting the moss uptake when the material is exposed in
bags. Moreover, the exposure was carried out in three European coun-tries
already evidenced in Hypnum cupressiforme and Pseudevernia fur-furacea differing in climate, meteorology, and in sites belonging to diverse land use
(Giordano et al., 2009). classes, in order to test the uptake effectiveness in different environments.
As reported in paragraph 2.2.5 the test on exposure at different heights
took place only in the urban and industrial sites. In both scenarios Cr, Fe, Hg
and Ni never met the 60% criterion we adopted, as well as Al in urban sites. The results evidenced that among the shape and the mesh size of the bags
In the case of Zn only the 4 m option tested no one clearly enhance the uptake capacity of the
372 F. Capozzi et al. / Environmental Pollution 214 (2016) 362e373
enclosed moss, as well as no differences were found among the moss exposed
at the heights tested. The aspects that affect more the element uptake are
represented by the density of the moss inside the bags and the relative ratio
between its weight and the surface area of the bag. In fact, in our test we
confirmed that the lower the density, the higher the uptake recorded.
Moreover, three weeks of exposure were not enough to have a consistent
uptake signal in the majority of the sites, while only a small difference does
occur be-tween 6 and 12 weeks of exposure. The above results hold true in all
exposure sites, regardless the different climatic conditions and land use
classes. Hence, comparisons among different bio-monitoring surveys should
be made only considering the data ob-tained with the same “experimental
settings” of those variables most affecting the moss uptake (i.e., species,
weight/surface ratio and exposure duration).
According to the reported outcomes, we suggest the use of a Mossphere,
because it is reusable, not “home-made” and with a regular and fixed shape; it
should be prepared with 1 or 2 mm mesh net (to avoid the loss of material)
and a moss content allowing a weight/surface ratio not higher than 15 mg cm
2
and it should be exposed at 4 m above the ground (this for practical reasons)
for a period not shorter than 6 weeks.
To further reduce variability in the samples, to improve the sensitivity of
the methodology and to apply sustainability stan-dards, the use of cloned
moss grown in vitro (Beike et al., 2015; Gonzalez et al., 2016a) is
recommended, as this material is even suitable to accumulate polycyclic
aromatic hydrocarbons (PAHs) (Concha-Grana et al., 2015). Furthermore, the
moss material should be devitalized like in our current study, because
Gonzalez et al. (2016b), demonstrated the metabolic activity of living moss
ma-terial for copper recently.
We support the adoption of a shared exposure protocol by the research
community, considering it the key aspect to make bio-monitoring surveys
directly comparable, also in view of its recog-nition as a monitoring method
by the EU legislation.
Acknowledgments
This study received financial support from FP7-ENV.2011.3.1.9e1
MOSSclone project of the European Union in the Seventh Framework
Program (FP7) for Research and Technological Development. Thanks are also
due to: Stadt Wien - Magis-tratsabteilung 33 & Magistratsabteilung 49, Stadt
Linz e
Gebaudemanagement€ und Tiefbau, Stadtwerke Leoben,
€ €
Osterreichische Bundesforste (OBF), Familie Schrammel, Michael Andert,
Veronika Fasching.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.envpol.2016.04.043.
References
Adamo, P., Giordano, S., Vingiani, S., Cobianchi, R.C., Violante, P., 2003. Trace element
accumulation by moss and lichen exposed in bags in the city of Naples (Italy). Environ.
Pollut. 122, 91e103.
Adamo, P., Giordano, S., Minganti, V., Modenesi, P., Monaci, F., Pittao, E., Tretiach, M.,
Bargagli, R., 2007. Lichen and moss bags as monitoring device in urban areas. Part II: trace
element content in living and dead biomonitors and comparison with synthetic materials.
Environ. Pollut. 146, 392e399.
Adamo, P., Giordano, S., Sforza, A., Bargagli, R., 2011. Implementation of airborne trace
element monitoring with devitalised transplants of Hypnum cupressiforme Hedw.:
assessment of temporal trends and element contribution by vehicular traffic in Naples city.
Environ. Pollut. 159, 1620e1628.
Anicic, M., Tasic, M., Frontasyeva, M.V., Tomasevic, M., Rajsic, S., Strelkova, L.P., Popovic,
A., Steinnes, E., 2008. Active biomonitoring with wet and dry moss: a case study in an
urban area. Environ. Chem. Lett. 7, 55e60.
Anicic, M., Tomasevic, M., Tasic, M., Rajsic, S., Popovic, A., Frontasyeva, M.V., et al., 2009.
Monitoring of trace element atmospheric deposition using dry and wet moss bags:
accumulation capacity versus exposure time. J. Hazard. Mater 171, 182e188.
Archibold, O.W., 1985. The metal content of wind-blown dust from uranium tailings in
Northern Saskatchewan. Water Air Soil Pollut. 24, 63e76.
Ares, A., Aboal, J.R., Carballeira, A., Giordano, S., Adamo, P., Fernandez, J.A., 2012. Moss bag
biomonitoring: a methodological review. Sci. Total Environ. 432, 143e158.
Ares, A., Fernandez, J.A., Carballeira, A., Aboal, J.R., 2014. Towards the methodo- logical
optimization of the moss bag technique in terms of contaminants con- centrations and
replicability values. Atmos. Environ. 94, 496e507.
Ares, A., Aboal, J.R., Carballeira, A., Fernandez, J.A., 2015. Do moss bags containing
devitalized Sphagnum denticulatum reflect heavy metal concentrations in bulk deposition?
Ecol. Indic. 50, 90e98.
Basile, A., Sorbo, S., Aprile, G., Conte, B., Cobianchi, R.C., 2008. Comparison of the heavy
metal bioaccumulation capacity of an epiphytic moss and an epiphytic lichen. Environ.
Pollut. 151, 401e407.
Basile, A., Sorbo, S., Aprile, G., Conte, B., Castaldo Cobianchi, R., Pisani, T., Loppi, S., 2009.
Heavy metal deposition in the Italian “triangle of death ” determined with the moss
Scorpiurum circinatum. Environ. Pollut. 157, 2255e2260.
Beelen, R., et al., 2014. Effects of long-term exposure to air pollution on natural- cause
mortality: an analysis of 22 European cohorts within the multicentre ESCAPE project.
Lancet 383, 785e795.
Beike, A.K., Spagnuolo, V., Lüth, V., Steinhart, F., Ramos-Gomez, J., Krebs, M., Adamo, P.,
Rey-Asensio, A.I., Giordano, S., Decker, E.L., Reski, R., 2015. Clonal in vitro propagation
of peat mosses (Sphagnum L.) as novel Green resources for basic and applied research.
PTOC 120, 1037e1049.
Boquete, T., Fernandez, J.A., Carballeira, A., Aboal, J.R., 2013. Assessing the tolerance of the
terrestrial moss Pseudoscleropodium purum to high levels of atmospheric heavy metals: a
reciprocal transplant study. Sci. Total Environ. 461e462, 552e559.
Concha-Grana, E., Muniategui-Lorenzo, S., De Nicola, F., Aboal, J.R., Rey-Asensio, A.I.,
Giordano, S., Reski, R., Lopez-Mahia, P., Prada-Rodriguez, D., 2015. Matrix solid phase
dispersion method for determination of polycyclic aromatic hydrocar- bons in moss. J.
Chromatogr. A 1406, 19e26.
Couto, J.A., Aboal, J.R., Fernandez, J.A., Carballeira, A., 2004. A new method for testing the
sensitivity of active biomonitoring: an example of its application to a terrestrial moss.
Chemosphere 57, 303e308.
De Nicola, F., Murena, F., Costagliola, M.A., Alfani, A., Baldantoni, D., Prati, M.V., Sessa, L.,
Spagnuolo, V., Giordano, S., 2013. A multi-approach monitoring of particulate matter,
metals and PAHs in an urban street canyon. Environ. Sci. Pollut. Res. 20, 4969e4979.
EEA Report No 4/2012. Air Quality in Europe d 2012 Report. ISSN 1725e9177. EEA Report
No 9/2013. Air Quality in Europe d 2013 Report. ISSN 1725e9177. Gailey, F.A.Y., Lloyd, O.L.,
1986a. Atmospheric metal pollution monitored by spher-
ical moss bags: a case study of Armadale. Environ. Health Perspect. 68, 187e196.
Gailey, F.A.Y., Lloyd, O.L., 1986b. Methodological investigations into low technology
monitoring of atmospheric metal pollution: part I. The effects of sampler size on metal
concentrations. Environ. Pollut. Ser. B Chem. Phys. 12, 41e59.
Gailey, F.A.Y., Lloyd, O.L., 1986c. Methodological investigations into low technology
monitoring of atmospheric metal pollution: part II. The effects of length of exposure on
metal concentrations. Environ. Pollut. Ser. B Chem. Phys. 12, 61e74.
Gailey, F.A.Y., Lloyd, O.L., 1986d. Methodological investigations into low technology
monitoring of atmospheric metal pollution: part III. The degree of replicability of the metal
concentrations. Environ. Pollut. Ser. B Chem. Phys. 12, 85e109.
Giordano, S., Adamo, P., Monaci, F., Pittao, E., Tretiach, M., Bargagli, R., 2009. Bags with
oven-dried moss for the active monitoring of airborne trace elements in urban areas.
Environ. Pollut. 157, 2798e2805.
Giordano, S., Adamo, P., Spagnuolo, V., Tretiach, M., Bargagli, R., 2013. Accumulation of
airborne trace elements in mosses, lichens and synthetic materials exposed at urban
monitoring stations: towards a harmonisation of the moss-bag tech- nique. Chemosphere 90,
292e299.
Gonzalez, A.G., Pokrovsky, O.S., Beike, A.K., Reski, R., Di Palma, A., Adamo, P., Giordano,
S., Fernandez, J.A., 2016a. Metal and proton adsorption capacities of natural and cloned
Sphagnum mosses. J. Colloid Interface Sci. 461, 326e334.
Gonzalez, A.G., Jimenez-Villacorta, F., Beike, A.K., Reski, R., Adamo, P., Pokrovsky, O.S.,
2016b. Chemical and structural characterization of copper adsorbed on mosses
(Bryophyta). J. Hazard. Mater 308, 343e354.
Goodman, G.T., Roberts, T.M., 1971. Plants and soils as indicators of metals in the air.
Nat. (Lond.) 231e287.
Harmens, H., Mills, G., Hayes, F., Norris, D.A., Sharps, K., 2015. Twenty-eight years of ICP
Vegetation: an overview of its activities. Ann. Bot. 5, 31e43.
Laschober, C., Limbeck, A., Rendl, J., Puxbaum, H., 2004. Particulate emissions from on-road
vehicles in the Kaisermühlen-tunnel (Vienna, Austria). Atmos. Environ. 38, 2187e2195.
Little, P., Martin, M.H., 1974. Biological monitoring of heavy metal pollution. Envi- ron. Pollut.
6, 1e19.
Napier, F., D'Arcy, B., Jefferies, C., 2008. A review of vehicle related metals and polycyclic
aromatic hydrocarbons in the UK environment. Desalination 226, 143e150.
Ratcliffe, J.M., 1975. An evaluation of the use of biological indicators in an atmo- spheric lead
survey. Atmos. Environ. 9, 623e629.
Rivera, M., Zechmeister, H., Medina-Ramon, M., Basagana,~ X., Foraster, M., Bouso, L.,
 F. Capozzi et al. / Environmental Pollution 214 (2016) 362e373
Moreno, T., Solanas, P., Ramos, R., Kollensperger,€ G., Deltell, A., Vizcaya, D., Künzli, N.,
2011. Monitoring of heavy metal concentrations in home outdoor air using moss bags.
Environ. Pollut. 159, 954e962.
Steinnes, E., Rühling, Å., Lippo, H., Makinen, A., 1997. Reference materials for large- scale
metal deposition surveys. Accred. Qual. Assur 2, 243e249.
Strachan, W.M.J., Glooschenko, W.A., 1988. Moss bags as monitors of organic contamination in
the atmosphere. Bull. Environ. Contam. Toxicol. 40, 447e450.
Tavares, H.M.C.F., Vasconcelos, M.T.S.D., 1996. Comparison of lead levels collected by
Sphagnum auriculatum and by a low-volume aerosol sampler in the urban at- mosphere of
Oporto. Toxicol. Environ. Chem. 54, 195e209.
Temple, P.J., McLaughlin, D.L., Linzon, S.N., Wills, R., 1981. Moss bags as monitors of
atmospheric deposition. J. Air Pollut. Control. Assoc. 31, 668e670.
Thorpe, A., Harrison, R.M., 2008. Sources and properties of non-exhaust particulate matter from
road traffic: a review. Sci. Total Environ. 400, 270e282.
Vasconcelos, M.T.S.D., Tavares, H.M.F., 1998. Atmospheric metal pollution (Cr, Cu, Fe, Mn,
Ni, Pb and Zn) in Oporto city derived from results for low-volume aerosol samplers and for
the moss Sphagnum auriculatum bioindicator. Sci. Total. En-viron. 212, 11e20.
View publication stats
373
Vukovic, G., Anicic Urosevic, M., Razumenic, I., Goryainova, Z., Frontasyeva, M., Tomasevic,
M., Popovic, A., 2013. Active moss biomonitoring of small-scale spatial distribution of
airborne major and trace elements in the Belgrade ur- ban area. Environ. Sci. Pollut. Res. 20,
5461e5470.
Zechmeister, H.G., Hohenwallner, D., Riss, A., Hanus-Illnar, A., 2005. Estimation of element
deposition deriving from road traffic sources by mosses. Environ. Pollut. 138, 238e249.
Zechmeister, H.G., Hagendorfer, H., Hohenwallner, D., Hanus-Illnar, A., Riss, A., 2006.
Analyses of platinum group elements in mosses as indicators of road traf fic emissions in
Austria. Atmos. Environ. 40, 7720e7732.
Sitography
http://www.zamg.ac.at/.
http://www.sito.regione.campania.it/.
http://www.meteogalicia.es/.